生成式人工智能赋能地方高校动力工程专业学位硕士培养模式改革研究

Reform on Generative-AI-Empowered Training Mode of Professional Master in Power Engineering for Local Universities

摘要: 地方高校动力工程专业学位硕士存在生源异质性突出、学科壁垒明显、实践教学资源短缺、评价重终结轻过程等培养难题。立足地方高校办学定位，紧扣工程教育认证理念，本文依托生成式人工智能技术探索专硕培养改革路径，从CFD私有知识库、多学科交叉培养、AI阶梯式实践训练、三维过程化评价、师生–AI协同闭环教学五个方面构建改革方案，落实交叉课程、CFD智能插件、虚拟仿真实验库等教改举措。试点结果显示，新模式可降低仿真编程门槛，打破学科壁垒，实现从结果导向向过程赋能转变，提升研究生工程实践能力与科研动力，可为同类地方高校工科专硕改革提供理论参考与实践范式。

Abstract: Local universities are facing multiple challenges in training professional masters in power engineering, such as prominent heterogeneity of student backgrounds, distinct disciplinary barriers, insufficient practical teaching resources, and an evaluation system that prioritizes summative assessment over process-oriented development. Grounded in the educational positioning of local universities and guided by the philosophy of engineering education accreditation, this paper explores reform pathways for professional master training enabled by generative artificial intelligence. A reform framework is constructed covering five dimensions: a private CFD knowledge base, interdisciplinary training, AI-driven staged practical training, three-dimensional process-oriented evaluation, and a closed-loop teaching system of teacher-student-AI collaboration. Multiple teaching-reform measures including interdisciplinary courses, intelligent CFD plugins and a virtual simulation experiment library have been implemented. Pilot results show that the proposed model can lower barriers to simulation programming, break down disciplinary boundaries, shift the training paradigm from outcome-orientation to process-driven development, and enhance postgraduates’ practical engineering competence and research motivation. It can provide theoretical references and practical paradigms for the reform of engineering professional masters at similar local universities.